Double-mechanical-arm stability control method for compensating gap characteristic
Through the state sensor and environmental sensor combining a collaborative control model and neural network algorithm, the state of the dual robotic arms is regulated in real time, solving the problem of poor collaborative control performance of the dual robotic arms in a port environment, and achieving more efficient synchronous and stable operation.
Patent Information
- Application Number
- CN202410851824.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-06-28
- Publication Date
- 2024-11-01
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
During the cargo handling process of the port, due to the large differences between the external environment and the indoor environment, the coordinated control performance is deteriorated, making it difficult to respond to changes in the external environment in a timely manner, affecting the handling effect.
The state sensor and environmental sensor are used to obtain the control status and external environment data of the dual robot arms in real time, and data analysis and regulation are performed through the collaborative control model and the optimization neural network algorithm to achieve synchronous and stable operation of the dual robot arms.
It improves the coordinated control performance of the dual robotic arms, can respond to changes in the external environment in a timely manner, enhances the synchronization and stability of operations, and improves work efficiency and flexibility.
Smart Images

Figure CN118876033A_ABST
Abstract
Description
Technical Field
[0001] The invention belongs to the technical field of mechanical arm control, and in particular relates to a dual mechanical arm stability control method for compensating for gap characteristics. Background Art
[0002] The dual manipulator is a serial robot composed of multiple joints. It has two independent manipulator structures. Each manipulator contains multiple movable joints to simulate the movement of human arms. The dual manipulator consists of two independent manipulators. Each manipulator contains fixed joints and movable joints. These joints can move on multiple basic axes, such as left-right axis, front-back axis and up-down axis, to achieve complex operation tasks. The dual manipulator can complete many complex operation tasks, such as carrying large objects, precise operation, and surgical assistance. Because it has two independent manipulators, it can perform different operations at the same time, which greatly improves work efficiency and flexibility. In the process of cargo handling at the port, the dual manipulator needs to use two independent controllers to control the motion during the cooperation of the dual manipulators, because the external environment is much different from the indoor working environment. The coordination performance between them becomes worse. In order to improve the handling effect of the dual manipulators, the coordinated control of the dual manipulators will be more complicated, inconvenient to control, difficult to respond in time, and difficult to make synchronous adjustments with the external environment. In this regard, we propose a dual manipulator stability control method that compensates for the gap characteristics. Summary of the invention
[0003] The technical problem to be solved by the present invention is to provide a dual-manipulator stability control method for compensating for gap characteristics.
[0004] The technical solution adopted by the present invention to solve the above technical problem is: a dual-manipulator stability control method for compensating gap characteristics, the control steps are:
[0005] S1. Develop and design a collaborative control model, substitute the data set into the designed collaborative control model, and adjust the model parameters to optimize the model;
[0006] S2, writing the model into the control system of the robot arm to perform operation control;
[0007] S3. Use the state sensor to obtain the control state data of the dual robotic arms in real time, upload the data to the collaborative control model, and make timely corrections and synchronization of collaborative operations;
[0008] S4. Obtain external environmental data through environmental sensors, upload them, analyze and identify the data through optimized neural network algorithms, obtain the external environmental conditions, upload the data to the collaborative control model, and regulate the dual robotic arms in combination with the external environment;
[0009] S5, control the dual robotic arms to make timely corrections and control the dual robotic arms to operate synchronously and stably;
[0010] The collaborative control model obtains the working status of the dual robotic arms based on state sensors and environmental sensors, and judges and controls the collaborative work of the dual robotic arms. The environmental sensors collect external environmental data, analyze and identify the external environment through optimized neural network algorithms, and upload the working environment data to the collaborative control model. The state sensors collect the working status of the dual robotic arms and upload it to the collaborative control model.
[0011] Preferably, the collaborative control model design step in step S1 defines a state equation, which is expressed as: x=f(x, d, t), where x is the state variable of the system, d is the control variable, and t is time, to set the control target of the model, the control target is the working condition of the dual robotic arms, and define a macro variable p(x, t), which is a function of the system state variable x, expressed as x1-x n A collaborative control model is designed based on the linear combination of the dataset, and the data set is substituted into the designed collaborative control model to optimize the parameters and improve the model.
[0012] Preferably, the function of the system state variable x is a set of first-order differential equations that describe the law of change of the state variable, and the system state formula is:
[0013] x=Ax+Bu
[0014] Where x is the state variable, A is the system matrix, B is the input matrix, and u is the system input. The formula describes the internal state of the system evolving over time.
[0015] Preferably, in step S2, the model is written by programming input in the control system of the robot arm, and the code of the model is input into the control system of the robot arm to perform test operation control.
[0016] Preferably, in step S3, the state sensor includes a visual sensor, a force sensor, a position sensor and an inertial measurement unit. The visual sensor is used to obtain the working environment image of the dual robotic arms, the force sensor is used to measure the contact force between the robotic arms and the working object, the position sensor is used to measure the position and angle of each joint of the robotic arms, and the inertial measurement unit includes an accelerometer and a gyroscope to measure the acceleration and angular velocity of the robotic arms. By uploading the data measured by the state sensor to the collaborative control model, the control system can perform operational control.
[0017] Preferably, the environmental sensor in step S4 includes a humidity sensor, a temperature sensor and a wind speed sensor, wherein the humidity sensor is used to measure the humidity in the working environment, the temperature sensor is used to measure the temperature in the working environment, and the wind speed sensor is used to measure the wind speed in the environment. The parameter conditions in the working environment are obtained through the environmental sensors, and the parameter data of the working environment is uploaded.
[0018] Preferably, the neural network algorithm optimization in step S4 is optimized by stochastic gradient descent method. The optimization step initializes the weights and biases of the neural network, selects a sample from the training set as the parameter update in the iteration, inputs the selected sample features into the neural network, obtains the predicted output of the network, calculates the error, and back propagates according to the calculated error to update the parameters of the model; inputs the data collected by the external sensor into the optimized neural network algorithm to obtain the working environment conditions of the dual robotic arms, and inputs the working environment conditions into the collaborative control model.
[0019] Preferably, the state equation of the collaborative control model is the state equation of the dual robotic arms, and the state working conditions of the dual robotic arms include the error of collaborative motion and the error of the environment. The error of collaborative motion refers to the inaccurate time synchronization between the robotic arms during the collaborative motion process, which leads to collaborative error. The environmental error refers to the error caused by the robotic arms being disturbed by temperature, humidity and wind speed in a complex environment.
[0020] Preferably, the coordinated control model achieves optimal control of the collaborative control model through differential optimization, and uses filtering to smooth the differential term, which responds to the changing trend of the deviation and performs advance adjustment.
[0021] Preferably, the coordinated control model adjusts the control parameters of the dual robotic arms in real time according to the uploaded environmental data.
[0022] The beneficial effects of the present invention are as follows: the present application adopts state sensors and environmental sensors to obtain the real-time state of the dual robotic arms and the external working environment, and adjusts the state parameters of the dual robotic arms in real time according to the external environment, so as to better control the coordination of the dual robotic arms, respond in time, and make synchronous adjustments with the corresponding operational changes of the external environment, bringing better prospects for use. BRIEF DESCRIPTION OF THE DRAWINGS
[0023] Figure 1 is a flow chart of the method of the present invention. DETAILED DESCRIPTION
[0024] The present invention will now be described in further detail in conjunction with the accompanying drawings and preferred embodiments. These drawings are simplified schematic diagrams, which only illustrate the basic structure of the present invention in a schematic manner, and therefore only show the components related to the present invention.
[0025] like Figure 1 As shown, the present invention provides a technical solution: a dual-manipulator stability control method for compensating gap characteristics, the control steps are:
[0026] S1. Develop and design a collaborative control model, substitute the data set into the designed collaborative control model, and adjust the model parameters to optimize the model;
[0027] S2, writing the model into the control system of the robot arm to perform operation control;
[0028] S3. Use the state sensor to obtain the control state data of the dual robotic arms in real time, upload the data to the collaborative control model, and make timely corrections and synchronization of collaborative operations;
[0029] S4. Obtain external environmental data through environmental sensors, upload them, analyze and identify the data through optimized neural network algorithms, obtain the external environmental conditions, upload the data to the collaborative control model, and regulate the dual robotic arms in combination with the external environment;
[0030] S5, control the dual robotic arms to make timely corrections and control the dual robotic arms to operate synchronously and stably;
[0031] The collaborative control model obtains the working conditions of the dual robotic arms based on the state sensor and the environmental sensor, and judges and controls the collaborative work of the dual robotic arms. The environmental sensor collects external environment data, analyzes and identifies the external environment by optimizing the neural network algorithm, and uploads the working environment data to the collaborative control model. The state sensor collects the working status of the dual robotic arms and uploads it to the collaborative control model.
[0032] It can make real-time adjustments to the state parameters of the dual robotic arms according to the external environment, to better control the coordination of the dual robotic arms, to respond in a timely manner, and to make synchronous adjustments to corresponding operational changes with the external environment.
[0033] Furthermore, in step S1, the collaborative control model design step defines the state equation, which is expressed as: x = f (x, d, t), where x is the state variable of the system, d is the control variable, and t is the time, to set the control target of the model. The control target is the working condition of the dual robotic arms, and the macro variable p (x, t) is defined, which is a function of the system state variable x, expressed as x1-x n The linear combination of the collaborative control model is designed, and the data set is substituted into the designed collaborative control model to optimize the parameters and improve the model;
[0034] The collaborative control model can achieve more stable and reliable system performance and more efficient task execution by integrating the behaviors of multiple control units or intelligent agents. The collaborative control model has strong adaptability and can achieve effective collaborative operations in different environments and conditions. In a complex and changing environment, the collaborative control model can maintain close cooperation and coordination between control units or intelligent agents through intelligent decision-making and adaptive adjustment.
[0035] Furthermore, the function of the system state variable x is a set of first-order differential equations that describe the law of change of the state variable. The system state formula is:
[0036] x=Ax+Bu
[0037] Where x is the state variable, A is the system matrix, B is the input matrix, u is the system input, and the formula describes the internal state of the system evolving over time;
[0038] The specific steps of building the collaborative control model are as follows:
[0039] Determine the state variables x of the system, where x1 to x n Represent different dimensions or states of the system, clarify the control target of the system, that is, the dual robotic arms controlled by the system, establish the system state equation, set the macro variable ψ(x, t) as a linear combination of the system state variables to characterize the overall behavior of the system, define the system manifold ψ(x) = 0 as the stable state or target state that the system expects to achieve, and design a collaborative controller through the state space method, combining the definition of macro variables and manifolds, so that the system can gradually converge to the manifold. Use the equation Tψ+ψ=0 to adjust the controller parameters, where T is a time constant that affects the convergence speed of the system. According to the design of the collaborative controller, implement the specific control logic, including data acquisition, processing, decision-making and output links, ensure that the control logic can respond to changes in the system state in real time, and adjust the control variables to maintain the stability of the system or reach the expected state. Verify the effectiveness of the collaborative control model in a simulation environment, test the performance of the model by simulating different scenarios and conditions, test it in an actual system, collect data and evaluate the real-time performance and robustness of the model, and optimize and adjust the collaborative control model according to the results of verification and testing to improve its performance and stability.
[0040] Furthermore, in step S2, the model is written by programming input in the control system of the robot arm, and the code of the model is input into the control system of the robot arm to perform experimental operation control.
[0041] Furthermore, in step S3, the state sensors include vision sensors, force sensors, position sensors and inertial measurement units. The vision sensors are used to obtain images of the working environment of the dual robotic arms. The force sensors are used to measure the contact force between the robotic arms and the working objects. The position sensors are used to measure the positions and angles of the joints of the robotic arms. The inertial measurement unit includes an accelerometer and a gyroscope to measure the acceleration and angular velocity of the robotic arms. By uploading the data measured by the state sensors to the collaborative control model, the control system can perform operational control.
[0042] Furthermore, in step S4, the environmental sensors include a humidity sensor, a temperature sensor and a wind speed sensor, wherein the humidity sensor is used to measure the humidity in the working environment, the temperature sensor is used to measure the temperature in the working environment, and the wind speed sensor is used to measure the wind speed in the environment. The parameter conditions in the working environment are obtained through the environmental sensors, and the parameter data of the working environment is uploaded.
[0043] Furthermore, in step S4, the neural network algorithm is optimized by stochastic gradient descent method. The optimization step initializes the weight and bias of the neural network, selects a sample from the training set as the parameter update in the iteration, inputs the selected sample features into the neural network, obtains the predicted output of the network, calculates the error, and back propagates according to the calculated error to update the parameters of the model; inputs the data collected by the external sensor into the optimized neural network algorithm to obtain the working environment of the dual robot arm, and inputs the working environment into the collaborative control model;
[0044] The optimization of the neural network algorithm can improve the efficiency and speed of training, enhance the performance of the model, and enhance the robustness and stability of the model, so as to better enable the dual robotic arms to be stably controlled and operated and reduce the computing resource effect.
[0045] Furthermore, the state equation of the collaborative control model is the state equation of the dual robotic arms. The state working conditions of the dual robotic arms include the error of collaborative motion and the error of the environment. The error of collaborative motion refers to the inaccurate time synchronization between the robotic arms during the collaborative motion process, which leads to collaborative error. The environmental error refers to the error caused by the robotic arms being disturbed by temperature, humidity and wind speed in a complex environment.
[0046] Detailed information on the working status of the dual robotic arms, especially specific parameters and numbers such as working range, load capacity, and accuracy, will vary depending on different robotic arm models and application scenarios. The robotic arms have different position accuracy and attitude accuracy to adapt to different task requirements. In addition, the specific working status of the robotic arms may also be affected by the control system, sensors, actuator hardware, and software.
[0047] Furthermore, the coordinated control model achieves optimal control of the collaborative control model through differential optimization, by using filtering to smooth the differential term, which responds to the changing trend of the deviation and performs advance regulation;
[0048] Differential optimization relies on the derivative information of the objective function. By calculating the derivative, we can understand the rate of change of the function at any point, and thus decide how to adjust the parameters to optimize the function. Differential optimization is usually an iterative process that approaches the optimal solution by continuously updating the parameters. Differential optimization can be applied to various types of objective functions, including linear functions, nonlinear functions, convex functions, and non-convex functions.
[0049] Furthermore, the coordinated control model adjusts the control parameters of the dual robotic arms in real time according to the uploaded environmental data.
[0050] The above description only describes the specific implementation mode of the present invention. Various examples do not limit the essential content of the present invention. After reading the description, ordinary technicians in the relevant technical field can modify or deform the specific implementation mode described above without departing from the essence and scope of the invention.
Claims
1. A dual-manipulator stability control method for compensating gap characteristics, characterized in that: The control steps are: S1. Develop and design a collaborative control model, substitute the data set into the designed collaborative control model, and adjust the model parameters to optimize the model; S2, writing the model into the control system of the robot arm to perform operation control; S3, using state sensors to obtain the control state data of the dual robotic arms in real time, uploading the data to the collaborative control model, and making timely corrections and synchronizations for collaborative operations; S4. Obtain external environmental data through environmental sensors, upload them, analyze and identify the data through optimized neural network algorithms, obtain the external environmental conditions, upload the data to the collaborative control model, and regulate the dual robotic arms in combination with the external environment; S5, control the dual robotic arms to make timely corrections and control the dual robotic arms to operate synchronously and stably; The collaborative control model obtains the working status of the dual robotic arms based on state sensors and environmental sensors, and judges and controls the collaborative work of the dual robotic arms. The environmental sensors collect external environmental data, analyze and identify the external environment through optimized neural network algorithms, and upload the working environment data to the collaborative control model. The state sensors collect the working status of the dual robotic arms and upload it to the collaborative control model.
2. The dual-manipulator stability control method for compensating gap characteristics according to claim 1 is characterized in that: In step S1, the collaborative control model design step defines the state equation, which is expressed as: x = f (x, d, t), where x is the state variable of the system, d is the control variable, and t is time. The control target of the model is set, and the control target is the working condition of the dual robotic arms. The macro variable p (x, t) is defined as a function of the system state variable x, expressed as x1-x n A collaborative control model is designed based on the linear combination of the dataset, and the data set is substituted into the designed collaborative control model to optimize the parameters and improve the model.
3. The dual-manipulator stability control method for compensating gap characteristics according to claim 2 is characterized in that: The function of the system state variable x is a set of first-order differential equations that describe the law of change of the state variable. The system state formula is: x=Ax+Bu Where x is the state variable, A is the system matrix, B is the input matrix, and u is the system input. The formula describes the internal state of the system evolving over time.
4. The dual-manipulator stability control method for compensating gap characteristics according to claim 1, characterized in that: In step S2, the model is written by programming input in the control system of the robot arm, and the model code is input into the control system of the robot arm to perform experimental operation control.
5. The dual-manipulator stability control method for compensating gap characteristics according to claim 1, characterized in that: In step S3, the state sensors include vision sensors, force sensors, position sensors and inertial measurement units. The vision sensors are used to obtain the working environment image of the dual robotic arms, the force sensors are used to measure the contact force between the robotic arms and the working objects, the position sensors are used to measure the positions and angles of the joints of the robotic arms, and the inertial measurement unit includes an accelerometer and a gyroscope to measure the acceleration and angular velocity of the robotic arms. By uploading the data measured by the state sensors to the collaborative control model, the control system can perform operational control.
6. The dual-manipulator stability control method for compensating gap characteristics according to claim 1, characterized in that: In step S4, the environmental sensors include a humidity sensor, a temperature sensor and a wind speed sensor, wherein the humidity sensor is used to measure the humidity in the working environment, the temperature sensor is used to measure the temperature in the working environment, and the wind speed sensor is used to measure the wind speed in the environment. The environmental sensors are used to obtain the parameter conditions in the working environment, and the parameter data of the working environment is uploaded.
7. The dual-manipulator stability control method for compensating gap characteristics according to claim 6, characterized in that: In step S4, the neural network algorithm is optimized by stochastic gradient descent method. The optimization step initializes the weights and biases of the neural network, selects a sample from the training set as the parameter update in the iteration, inputs the selected sample features into the neural network, obtains the predicted output of the network, calculates the error, and back propagates based on the calculated error to update the parameters of the model; the data collected by the external sensors is input into the optimized neural network algorithm to obtain the working environment conditions of the dual robotic arms, and the working environment conditions are input into the collaborative control model.
8. The dual-manipulator stability control method for compensating gap characteristics according to claim 2, characterized in that: The state equation of the collaborative control model is the state equation of the dual robotic arms. The state working conditions of the dual robotic arms include the error of collaborative motion and the error of the environment. The error of collaborative motion refers to the inaccurate time synchronization between the robotic arms during the collaborative motion process, which leads to collaborative error. The environmental error refers to the error caused by the robotic arms being disturbed by temperature, humidity and wind speed in a complex environment.
9. The dual-manipulator stability control method for compensating clearance characteristics according to claim 1, characterized in that: The coordinated control model achieves optimal control of the collaborative control model through differential optimization. The differential term is smoothed by using filtering. The differential term responds to the changing trend of the deviation and performs advance adjustment.
10. The dual-manipulator stability control method for compensating gap characteristics according to claim 1, characterized in that: The coordinated control model adjusts the control parameters of the dual robotic arms in real time based on the uploaded environmental data.
